Target Spectral Signature Research Articles

A Representation-Learning-Based Graph and Generative Network for Hyperspectral Small Target Detection

Hyperspectral small target detection (HSTD) is a promising pixel-level detection task. However, due to the low contrast and imbalanced number between the target and the background spatially and the high dimensions spectrally, it is a challenging one. To address these issues, this work proposes a representation-learning-based graph and generative network for hyperspectral small target detection. The model builds a fusion network through frequency representation for HSTD, where the novel architecture incorporates irregular topological data and spatial–spectral features to improve its representation ability. Firstly, a Graph Convolutional Network (GCN) module better models the non-local topological relationship between samples to represent the hyperspectral scene’s underlying data structure. The mini-batch-training pattern of the GCN decreases the high computational cost of building an adjacency matrix for high-dimensional data sets. In parallel, the generative model enhances the differentiation reconstruction and the deep feature representation ability with respect to the target spectral signature. Finally, a fusion module compensates for the extracted different types of HS features and integrates their complementary merits for hyperspectral data interpretation while increasing the detection and background suppression capabilities. The performance of the proposed approach is evaluated using the average scores of AUCD,F, AUCF,τ, AUCBS, and AUCSNPR. The corresponding values are 0.99660, 0.00078, 0.99587, and 333.629, respectively. These results demonstrate the accuracy of the model in different evaluation metrics, with AUCD,F achieving the highest score, indicating strong detection performance across varying thresholds. Experiments on different hyperspectral data sets demonstrate the advantages of the proposed architecture.

Advancements in Remote Compressive Hyperspectral Imaging: Adaptive Sampling with Low-Rank Tensor Image Reconstruction

We advanced the practical development of compressive hyperspectral cameras for remote sensing scenarios with a design that simultaneously compresses and captures high-quality spectral information of a scene via configurable measurements. We built a prototype imaging system that is compatible with light-modulation devices that encode the incoming spectrum. The sensing approach enables a substantial reduction in the volume of data collected and transmitted, facilitating large-scale remote hyperspectral imaging. A main advantage of our sensing design is that it allows for adaptive sampling. When prior information of a survey region is available or gained, the modulation patterns can be re-programmed to efficiently sample and detect desired endmembers. Given target spectral signatures, we propose an optimization scheme that guides the encoding process. The approach severely reduces the number of required sampling patterns, with the ability to achieve image segmentation and correct distortions. Additionally, to decode the modulated data, we considered a novel reconstruction algorithm suited for large-scale images. The computational methodology leverages the multidimensional structure and redundant representation of hyperspectral images via the canonical polyadic decomposition of multiway arrays. Under realistic remote sensing scenarios, we demonstrated the efficiency of our approach with several data sets collected by our prototype camera and reconstructed by our low-rank tensor decoder.

A GLRT-Based Multi-Pixel Target Detector in Hyperspectral Imagery

In hyperspectral imagery, target detection algorithms are usually based on the spectral signature information. Due to the advance of the spatial resolution of hyperspectral sensors, the ground sample distance may be much smaller than the size of targets. As a result, targets often occupy multiple consecutive pixels, which are referred to as multi-pixel targets. In this paper, we investigate the target detection problem for multi-pixel targets in hyperspectral imagery, when the target spectral signature is known. Jointly exploiting the pixels occupied by a target of interest, we propose a multi-pixel target detector resorting to the generalized likelihood ratio test criterion. Closed-form expressions for the probabilities of the false alarm and detection are derived, which are verified using Monte Carlo simulations. Experimental results on four real hyperspectral datasets show that the proposed detector outperforms its counterparts.

Triple coding empowered FDMA-CDMA mode high-security CAOS camera.

For the first time, to the best of our knowledge, the hybrid triple coding empowered frequency division multiple access (FDMA)-code division multiple access (CDMA) mode of the coded access optical sensor (CAOS) camera is demonstrated. Compared to the independent FDMA and CDMA modes, the FDMA-CDMA mode has a novel high-security space-time-frequency triple signal encoding design for robust, faster, linear irradiance extraction at a moderately high dynamic range (HDR). Specifically, this hybrid mode simultaneously combines the linear HDR strength of the FDMA-mode fast Fourier transform (FFT) digital signal processing (DSP)-based spectrum analysis with the high signal-to-noise ratio (SNR) provided by the many simultaneous CAOS pixels' photodetection of the CDMA mode. In particular, the demonstrated FDMA-CDMA mode with P FDMA channels provides a P times faster camera operation versus the equivalent linear HDR frequency modulation (FM)-CDMA mode. Visible band imaging experiments using a digital-micromirror-device-based CAOS camera operating in its passive light mode demonstrates a P=4 channels FDMA-CDMA mode, illustrating high-quality image recovery of a calibrated 64 dB six-patch HDR target versus the CDMA and FM-CDMA CAOS modes, which limit dynamic range and speed, respectively. For the first time to our knowledge, we demonstrate the simultaneous dual image capture capability of the FDMA-CDMA mode using silicon (Si) and germanium (Ge) large-area point photodetectors, allowing the capture of the ultraviolet-near-infrared 350-1,800 nm full spectrum. The active FDMA-CDMA mode CAOS camera operation is also demonstrated using P=3 LED light sources, each with its unique optical spectral content driven by its independent FDMA frequency. This illuminated target spectral signature matched active CAOS mode allows simultaneous capture of P images without the use of P time-multiplexed slot operation tunable optical filters. Applications for such a FDMA-CDMA camera includes controlled light illumination food inspection to bright light exposure security systems.

Coincidence of the Rao Test, Wald Test and GLRT for anomaly detection in hyperspectral imagery

Anomaly detection methods are designed to detect targets (small anomalies) without a priori information on the target spectral signature. In this letter, we deal with the problem of anomaly detection for hyperspectral images based on the Gaussian model assuming that the background obeys a real-valued Gaussian multivariate distribution with unknown covariance matrix. This model is widely used in hyperspectral images. We derive the corresponding Rao and Wald tests, and show that both the two tests are equivalent to the generalized likelihood ratio test.

A Novel Change Detection Method for Multitemporal Hyperspectral Images Based on Binary Hyperspectral Change Vectors

Hyperspectral (HS) images provide a dense sampling of target spectral signatures. Thus, they can be used in a multitemporal framework to detect and discriminate between different kinds of fine spectral change effectively. However, due to the complexity of the problem and the limited amount of multitemporal images and reference data, only a few works in the literature addressed change detection (CD) in HS images. In this paper, we present a novel method for unsupervised multiple CD in multitemporal HS images based on a discrete representation of the change information. Differently from the state-of-the-art methods, which address the high dimensionality of the data using band reduction or selection techniques, in this paper, we focus our attention on the representation and exploitation of the change information present in each band. After a band-by-band pixel-based subtraction of the multitemporal images, we define the hyperspectral change vectors (HCVs). The change information in the HCVs is then simplified. To this end, the radiometric information of each band is separately analyzed to generate a quantized discrete representation of the HCVs. This discrete representation is explored by considering the hierarchical nature of the changes in HS images. A tree representation is defined and used to discriminate between different kinds of change. The proposed method has been tested on a simulated data set and two real multitemporal data sets acquired by the Hyperion sensor over agricultural areas. Experimental results confirm that the discrete representation of the change information is effective when used for unsupervised CD in multitemporal HS data.

Multitask Learning-Based Reliability Analysis for Hyperspectral Target Detection

Hyperspectral images contain abundant spectral information, which provide great potential for target detection. However, it also introduces a critical spectral variability problem for hyperspectral target detection, which makes the hyperspectral target detection much difficult than the classical spectral match issue. Many traditional detection methods have been proposed to deal with the spectral variability. However, these algorithms are still highly susceptible to the target spectral variability. The single input restriction and the inherent spectral characteristics mining problem are the main issues with these methods. The multitask learning (MTL) technique may have the potential to solve the above hyperspectral target detection issues since it can extract the inherent similarity and difference within multiple priori target spectra to learn a robust target spectral signature. This paper proposed a MTL-based reliability analysis method for hyperspectral target detection (MultiRely). This approach: 1) utilizes multiple priori target spectra to better represent the target spectral characteristics and construct multiple related detection tasks; 2) takes full advantage of the multitask learning technique to explore the spectral similarity and difference between multiple priori target spectra; 3) and applies the reliability analysis to obtain a reliable target spectrum in order to alleviate the target spectral variability. Experiments on two real hyperspectral datasets and one synthetic hyperspectral dataset illustrated the effectiveness of the proposed algorithm compared to the state-of-the-art detectors.

Band selection method for subpixel target detection using only the target reflectance signature.

While offering powerful capabilities, the high dimensionality of hyperspectral images can make information extraction a challenge. For that reason, dimension reduction is a common data processing step. For the purpose of subpixel target detection, band selection is a dimension reduction method that can optimize results as well as reduce computation costs. However, existing band selection methods that are used for subpixel target detection require background spectral reflectance signatures to compare with the target signatures. These methods work well and offer a distinct advantage over other dimension reduction methods such as principal component analysis or nonnegative matrix factorization, but only when the background information is available. In this study, we developed a method that selected bands using only the target spectral reflectance signature. We tested this method using a utility prediction model, validated the results with real images, then cross-validated the results with simulated images that were associated with perfect truth data. We studied the detection statistics for a range of bands selected using this method and compared it to the results obtained from three other band selection methods. The motivation for developing this method was to be able to reduce the number of bands prior to collection when background information was not available. For an adaptive spectral imaging system with a tunable sensor, we would be able to optimize detection for a specific target and save data handling costs associated with transmitting, storing, and disseminating the data for information extraction. This method was also simple enough to be computed using a small on-board CPU, and modify the bands' selection criteria as the target changed.

A Sparse Representation Method for <italic>a Priori</italic> Target Signature Optimization in Hyperspectral Target Detection

Many target detectors commonly utilize a single a priori target spectral signature as an input. However, the detection results are greatly affected by the quality of the a priori target spectral signature because the spectral variability phenomenon is universal and anisotropic in hyperspectral image data. This paper proposes a sparse representation-based method to generate an optimized target spectrum from limited target training samples, which is able to alleviate the impact of spectral variability on hyperspectral target detection. When lacking comprehensive knowledge about the target object of interest, an optimized representative target spectrum should be expected to be reconstructed by the hyperspectral data themselves in a sparse representation manner following the characteristics of the data structure and then be generated by a set of selected candidate pixels that contain the target signal with a varying status. With the optimized a priori target signature, the experimental results of the detection of different characteristics of objects with three different types of hyperspectral images confirm the effectiveness, robustness, and generality performance of the proposed method.

Identification of possible source areas of stone raw materials combining remote sensing and petrography

ABSTRACTUnderstanding the location and distribution of raw materials used in the production of prehistoric artefacts is a significant part of archaeological research that aims to understand the interregional interaction patterns in the past. The aim of this study is to explore the regional locations of the source rock utilized in the production of stone bowls, which were unearthed at the Neolithic (approximately 6500–5500 BC) site of Domuztepe (Kahramanmaraş-Turkey), via a combination of remote-sensing methods, petrographic and chemical analyses. To accomplish this task, the stone bowls were identified mineralogically, geochemically and spectrally, and then mapped with Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) sensors. According to the defined mineralogical composition, which is iron-rich chlorite, the target areas were selected among geologically potential areas that would bear similar source rocks in near vicinity and the target spectral signature was searched within these target areas. In order to overcome the problem of spectral similarity of chlorite group to some other minerals such as carbonate or epidote group minerals, band ratioing (BR) and feature-oriented principal component analysis (FOPCA) were used with an integrated approach and then their results were filtered according to the outcomes of the relative absorption band-depth (RBD) images. The areas with highest potentials were vectorized and then field checked. Mineralogical investigations on the collected field samples reveal that there is a mineralogical match between the source and target material. One group of stone bowls samples have similar geochemical signatures as the field samples having ultramafic origins. However, there is another group of stone bowls samples which are geochemically dissimilar to the first group of field and bowls samples. The data regarding the geochemical signatures of these two groups indicate a genetic relation between the sample sets. Therefore, it is concluded that the source rock of a major portion of the stone bowls unearthed at Domuztepe most probably originated from the near vicinity of the site.

Detecting Aircraft in Low-Resolution Multispectral Images: Specification of Relevant IR Wavelength Bands

We address the problem of detecting a stealth aircraft flying far away from an observer with limited visibility conditions using their multispectral signature. In such environment, the aircraft is a very low-contrast target, i.e., the target spectral signature may have a similar magnitude to the background clutter. Therefore, methods accounting only for the spectral features of the target, while leaving aside its spatial pattern, may either lead to poor detection statistics or high false alarm rate. We propose a new detection method which accounts for both spectral and spatial dispersions, by inferring level sets of the Mahalanobis transform of the multispectral image. This combines the approach of the well-known Reed Xiaoli (RX) detector with some elements of the level set methods for shapes analysis. This algorithm is in turn used to specify the wavelength bands which maximize an aircraft detection probability, for a given false alarm rate. This methodology is illustrated in a typical scenario, consisting of a daylight air-to-ground full-frontal attack by a generic combat aircraft flying at low altitude, over a database of 30 000 simulated multispectral infrared signature (IRS). The results emphasize that, in the context of aircraft detection, there is great interest in using multispectral IRS rather than integrated IRS, as long as the IR bands are well chosen.

Adjusted Spectral Matched Filter for Target Detection in Hyperspectral Imagery

Supervised target detection and anomaly detection are widely used in various applications, depending upon the availability of target spectral signature. Basically, they are based on a similar linear process, which makes them highly correlated. In this paper, we propose a novel adjusted spectral matched filter (ASMF) for hyperspectral target detection, which aims to effectively improve target detection performance with anomaly detection output. Specifically, a typical case is presented by using the Reed-Xiaoli (RX) anomaly detector to adjust the output of supervised constrained energy minimization (CEM) detector. The adjustment is appropriately controlled by a weighting parameter in different detection scenarios. Experiments were implemented by using both synthetic and real hyperspectral datasets. Compared to the traditional single detection method (e.g., CEM), the experimental results demonstrate that the proposed ASMF can effectively improve its performance by utilizing the result from an anomaly detector (e.g., RX), particularly in situations with a complex background or strong anomalies.

Spectral Similarity Measure Using Frequency Spectrum for Hyperspectral Image Classification

A novel spectral similarity measure approach, which is named spectral frequency spectrum difference (SFSD), is proposed for hyperspectral image classification based on the frequency spectrum of spectral signature using the Fourier transform. Many important characteristics of spectral signature can be clearly reflected in the frequency spectrum. Therefore, the spectral similarity is defined as the frequency spectrum's difference between the target and reference signatures. The frequency spectrum analysis in this study suggests that the magnitude values of the first few low-frequency components for spectral signature can effectively represent the spectral similarity. To balance the difference between the low- and high-frequency components, the frequency spectrum of the target spectral signature is taken as the normalized factor in the SFSD method. Next, the U.S. Geological Survey spectral data and two hyperspectral remote sensing images were employed as test data in our validation experiments. The new SFSD proposed here was compared with the leading approaches in terms of the spectral discriminability and classification accuracy. Results show that the SFSD exhibits a relatively better performance and has more robust applications for hyperspectral image classification.

An Automatic Robust Iteratively Reweighted Unstructured Detector for Hyperspectral Imagery

Unstructured background model-based detectors, which usually utilize a single a priori target spectral signature as the input, have been successfully applied in various hyperspectral target detection applications. However, the detection results are greatly affected by the quality of the a priori target spectral signature, as the spectral variability phenomenon is universal in hyperspectral image data. This paper proposes an iteratively reweighted method to generate an optimal target spectrum from limited target training spectra, which is able to alleviate the spectral variation. When the priori target spectral can only be chosen from the hyperspectral image, an optimal target spectrum can be iteratively generated by adjusting the pixel signals with varying weights. The experimental results with three different types of real hyperspectral images confirm the robust performance of the proposed method.

Classification of hyperspectral remote sensing images using frequency spectrum similarity

An algorithm of hyperspectral remote sensing images classification is proposed based on the frequency spectrum of spectral signature. The spectral signature of each pixel in the hyperspectral image is taken as a discrete signal, and the frequency spectrum is obtained using discrete Fourier transform. The discrepancy of frequency spectrum between ground objects’ spectral signatures is visible, thus the difference between frequency spectra of reference and target spectral signature is used to measure the spectral similarity. Canberra distance is introduced to increase the contribution from higher frequency components. Then, the number of harmonics involved in the proposed algorithm is determined after analyzing the frequency spectrum energy cumulative distribution function of ground object. In order to evaluate the performance of the proposed algorithm, two hyperspectral remote sensing images are adopted as experimental data. The proposed algorithm is compared with spectral angle mapper (SAM), spectral information divergence (SID) and Euclidean distance (ED) using the product accuracy, user accuracy, overall accuracy, average accuracy and Kappa coefficient. The results show that the proposed algorithm can be applied to hyperspectral image classification effectively.

Target detection using difference measured function based matched filter for hyperspectral imagery

Automatic target detection is a research focus in hyperspectral image processing field. Algorithms such as the matched filter (MF) and the adaptive coherence/cosine estimator (ACE) directly use the prior knowledge of target spectral signatures to detect targets. In this paper, we propose a difference measured function based matched filter (DFMF), which could include the famous algorithm MF as a special case. The DFMF uses a new measured function to build an objective function, and utilizes the gradient descent method to find an optimal projection vector. After finding the optimal projection vector, the interesting targets can be detected in the projection space. The experimental results demonstrate the proposed algorithm could detect interesting targets effectively and performs better than some other experimental algorithms.

Influence of band selection and target estimation error on the performance of the matched filter in hyperspectral imaging

The matched filter is a widely used detector in hyperspectral detection applications because of its simplicity and its efficiency in practical situations. We propose to estimate its performance with respect to the number of spectral bands. These spectral bands are selected thanks to a genetic algorithm in order to optimize the contrast between the target and the background in the detection plane. Our band selection method can be used to optimize not only the position but also the linewidth of the spectral bands. The optimized contrast always increases with the number of selected bands. However, in practical situations, the target spectral signature has to be estimated from the image. We show that in the presence of estimation error, the maximum number of bands may not always be the best choice in terms of detection performance.

Robust high-order matched filter for hyperspectral target detection

A robust high-order matched filter (RHMF) for automatic target detection in hyperspectral images is proposed. The classical detection methods mainly focus on second-order statistics and do not take intrinsic uncertainty or variability of target spectral signatures into account. For automatic target detection in a hyperspectral image, most interesting targets usually occur with low probabilities and small population and they generally cannot be described by second-order statistics. Also, one difficult point in target detection is the inherent variability in target spectral signatures. Under such circumstances, the RHMF algorithm uses high-order statistics, and takes variability into consideration, and has been shown by presented experiments to be more effective than classical detection methods.

A projection pursuit algorithm for anomaly detection in hyperspectral imagery

The main goal of this paper is to propose an innovative technique for anomaly detection in hyperspectral imageries. This technique allows anomalies to be identified whose signatures are spectrally distinct from their surroundings, without any a priori knowledge of the target spectral signature. It is based on an one-dimensional projection pursuit with the Legendre index as the measure of interest. The index optimization is performed with a simulated annealing over a simplex in order to bypass local optima which could be sub-optimal in certain cases. It is argued that the proposed technique could be considered as seeking a projection to depart from the normal distribution, and unfolding the outliers as a consequence. The algorithm is tested with AHS and HYDICE hyperspectral imageries, where the results show the benefits of the approach in detecting a great variety of objects whose spectral signatures have sufficient deviation from the background. The technique proves to be automatic in the sense that there is no need for parameter tuning, giving meaningful results in all cases. Even objects of sub-pixel size, which cannot be made out by the human naked eye in the original image, can be detected as anomalies. Furthermore, a comparison between the proposed approach and the popular RX technique is given. The former outperforms the latter demonstrating its ability to reduce the proportion of false alarms.

A Theoretical Framework for Hyperspectral Anomaly Detection Using Spectral and Spatial A Priori Information

This letter presents a new theoretical approach for anomaly detection using a priori information about targets. This a priori knowledge deals with the general spectral behavior and the spatial distribution of targets. In this letter, we consider subpixel and isolated targets that are spectrally anomalous in one region of the spectrum but not in another. This method is totally different from matched filters that suffer from a relative sensitivity to low errors in the target spectral signature. We incorporate the spectral a priori knowledge in a new detection distance, and we propose a Bayesian approach with a Markovian regularization to suppress the potential targets that do not respect the spatial a priori. The interest of the method is illustrated on simulated data consisting in realistic anomalies that are superimposed on a real HyMap hyperspectral image.